DislikedLike I said Oanda has that but they aren't an ECN and will probably have worse quotes.
IB comes close but they have a minimum that is too high.Ignored
Oanda is a bucketshop, I wouldn't use them.
Trade another day through Systematic Trading 203 replies
Nonparametric Association Measures in Systematic Trading 57 replies
Anyone use Systematic from Alpari? 1 reply
A Systematic Approach to Markets 68 replies
Systematic Trading vs. Discretionary Trading 14 replies
DislikedLike I said Oanda has that but they aren't an ECN and will probably have worse quotes.
IB comes close but they have a minimum that is too high.Ignored
DislikedThat's interesting - you don't do much trading system devlopment then? Or how you determine that your systems are profitable in the long term?Ignored
DislikedCraig have you ever compared the actual real time quote prices between IB and MB? As they are both ECN's they should be very similar. My thinking is, use IB's quotes and history but submit the orders to MB(could even do a check if MB's quote is within error margin). Best of both worlds (-:Ignored
DislikedThe second categories are systems which depend on tick data, these can't really be back tested anyway, do I just let it rip in paper mode and see what happens.Ignored
DislikedI am not sure if I agree with this. I collect every bid/ask tick and play it back. Ok it won't 100% and obviously there is slippage but it should be pretty close.Ignored
DislikedI haven't been able to come up with anything useful just based on the tick. I tried some stuff on keeping track of ticks as an indicator of voatility but nothing with any success. Have you managed any good systems based on tick data?Ignored
DislikedI'm trading about 3000 separate market/system combinations. I did not invent this, I got it from a guy on ET.Ignored
DislikedI am not sure if I agree with this. I collect every bid/ask tick and play it back. Ok it won't 100% and obviously there is slippage but it should be pretty close.Ignored
DislikedI read about acrary too a couple of months ago(thx to you guys). This is why I put this stuff into a table. Plan is to run the simulator on a load of strategies pick the top performers based on last days and use them on real account and off you go.Ignored
DislikedI paper trade systems across many markets in parallel, when they reach a certain measure of profitability they get automatically switched into live trading, I basically wait for the market to come to me. At the moment I'm trading about 3000 separate market/system combinations. I did not invent this, I got it from a guy on ET.
In the wider view I divide systems up into 2 categories, you have your old standards (breakout, trend etc) which have fairly well known implementations, I don't think that these need to be endlessly tweaked on historical data....Ignored
DislikedSorry file was done from unix, corrected it
ok tweeked the strategy a bit but basic idea is the same. I know the sample size is not right yet but the fact it is positive over 8 currencies gives me confidence. And I know most of these currencies are correlated with the dollar but I do question how much that really matters. I have found it really difficult to find strategies that are positive across 8 currencies so the fact that this one is should say something. In some respect I think you even though it is the same strategy you could see this as...Ignored
DislikedI would also really, really suggest using risk relative position sizing, otherwise you are basing your risk to volatility (assuming you are using volatility adjusted stops).Ignored
DislikedMikom,
Could you go into this a little bit more.
Do you mean number of units used based on the risk profile. How could you do this? Do you mean something like
if tscore>2 and profit_factor>2
then use 25000
else
use some type of ratio determing the position size
endIgnored